dleeftink

35 ms·

451.
by dleeftink 3y ago
Just to make sure I understand your excitement: we need guinea-pigs ahem people to wear 'head mounted all day ego centric AR' with who knows how many integrated sensors for long stretches on end, so we can finally get to our f
452.
by dleeftink 3y ago
Thank you for posting, I enjoyed the article. In true cybernetics fashion, the maximum compression ratio of information seems to be the current grail of our machine learning efforts, whether that is through inferring heuristics or graduall
453.
by dleeftink 3y ago
> but may have happened on occasion I would be interested to see the percentage of news items interfered/suppressed by third-parties vs stories that are not (specifically in the case of public broadcasters)[1]. As you are right to p
454.
by dleeftink 3y ago
Yes, the dichotomy falls apart quite easily. But in the context of the crudely worded title and the observations in the article that 'all news is propaganda', we at least need to look further then blaming 'the news' itse
455.
by dleeftink 3y ago
Which of the networks are you referring to specifically? Publically funded does not equate to newsroom not being able to operate independently (and is not necessarily the opposite to 'for-profit'). While completely transparent and
456.
by dleeftink 3y ago
I would prepend 'for-profit' to the title, but also make a distinction between the journalists's intent and affiliation when considering what news can be deemed independent or propaganda-like. That said, even for profit journ
457.
by dleeftink 3y ago
I think it is a cool 'tangible' demo of how non-human readable data remains learnable for ML purposes. I've recently tried using Nilsimsa and sdeep hashes to reduce the input dimensionality of natural text, with some loss in
458.
by dleeftink 3y ago
A mechanism that quite possibly exacerbates a lower degree of mutual understanding between the signing parties. It would be nice to have a 'summary' feature built-into the interface.
459.
by dleeftink 3y ago
True, but there's an marked difference between signing a traditional contract and clicking a ToS button as far as your upfront involvement is concerned. At least in my experience, there is much more 'back-and-forth' between
460.
by dleeftink 3y ago
As with many ML-phrases, the use of 'unlearning' frames data modelling as a false dichotomy between 'learnable' and 'forgettable' data. Whereas humans are able to forget over time, it would be quite disturbing
461.
by dleeftink 3y ago
If you/we/other folk want it to be a thing, shouldn't we at least try to not frame it as a 'dead in the water' piece of tech? Seeing how HN has a way of planting seeds in the minds of potential early adoptors, frami
462.
Burning of the Hooks: a close look at the current platform pyres(medium.com)
1 points by dleeftink 3y ago|0 comments